Regulation of DNA Methylation Using Different Tensions ofDouble Strands Constructed in a Defined DNA Nanostructure

Masayuki Endo,*,†,§ Yousuke Katsuda,‡ Kumi Hidaka,‡ and Hiroshi Sugiyama*,†,‡,§

Institute for Integrated Cell-Material Sciences (iCeMS), Kyoto UniVersity, Yoshida-ushinomiyacho,Sakyo-ku, Kyoto 606-8501, Japan, Department of Chemistry, Graduate School of Science, Kyoto

UniVersity, Kitashirakawa-oiwakecho, Sakyo-ku, Kyoto 606-8502, Japan, and CREST, Japan Scienceand Technology Corporation (JST), Sanbancho, Chiyoda-ku, Tokyo 102-0075, Japan

Received September 9, 2009; E-mail: endo@kuchem.kyoto-u.ac.jp; hs@kuchem.kyoto-u.ac.jp

Abstract: A novel strategy for regulation of an enzymatic DNA modification reaction has been developedby employing a designed nanoscale DNA scaffold. DNA modification using enzymes often requires bendingof specific DNA strands to facilitate the reaction. The DNA methylation enzyme EcoRI methyltransferase(M.EcoRI) bends double helix DNA by 55°-59° during the reaction with flipping out of the second adeninein the GAATTC sequence as the methyl transfer reaction proceeds. In this study, two different doublehelical tensions, tense and relaxed states of double helices, were created to control the methyl transferreaction of M.EcoRI and examine the structural effect on the methylation. We designed and prepared atwo-dimensional (2D) DNA scaffold named the “DNA frame” using the DNA origami method thataccommodates two different lengths of the double-strand DNA fragments, a tense 64mer double strandand a relaxed 74mer double strand. Fast-scanning atomic force microscope (AFM) imaging revealed thedifferent dynamic movement of the double-strand DNAs and complexes of M.EcoRI with 64mer and 74merdouble-strand DNAs. After treatment of the double strands in the DNA frame with M.EcoRI and thesubsequent digestion with restriction enzyme EcoRI (R.EcoRI), AFM analysis revealed that the 74merdouble-strand DNA was not effectively cleaved compared with the 64mer double-strand DNA, indicatingthat the methylation preferentially occurred in the relaxed 74mer double-strand DNA compared with that inthe tense 64mer double strand. Biochemical analysis of the methylation and specific digestion using areal-time PCR also supported the above results. These results indicate the importance of the structuralflexibility for bending of the duplex DNA during the methyl transfer reaction with M.EcoRI. Therefore, theDNA methylation can be regulated using the structurally controlled double-strand DNAs constructed in theDNA frame nanostructure.

Introduction

DNA-binding proteins including modifying enzymes first bindto nonspecific sites and diffuse along the DNA to search outspecific sequences and then modify the corresponding sites.1

DNA modification often needs structural changes at the targetDNA strands, such as DNA bending, for the reaction to proceed.Among the DNA modifications, DNA methylation plays acentral role in maintaining the biological system by protectingit from invasion by external genomes and epigenetic activation/deactivation of the transcription seen in the differentiation andreprogramming of the cell.2 DNA methyltransferases introducemethyl groups to the C5 and N4 positions of cytosine and theN6 position of adenine in the presence of S-adenosyl-L-methionine (SAM).2 EcoRI methyltransferase (M.EcoRI) bendsdouble helix DNA by 55°-59° with flipping out of the secondadenine in the GAATTC sequence as DNA methylationproceeds.3,4 DNA bending with M.EcoRI was observed at the

single-molecule level using DNA-gold nanoparticle conjugatescontaining a target sequence, which was monitored by measuringthe distance between two nanoparticles.5 From the structuralpoint of view, if the DNA bending is artificially controlled, DNAmethylation should be suppressed or promoted. For example,the artificial DNA bending by 30° using a tense alkyl linkerconnected by a disulfide linkage showed promotion of thebinding of HMG-D protein to duplex DNA.6 If the differenttensions of the double helices, tense and relaxed double-strandDNAs, are prepared, the activation or deactivation of the methyltransfer reactions for specific DNA strands can be controlledand examined.

For the preparation of structurally controlled double-strandDNAs, versatile scaffolds have been designed and prepared byutilizing the methods developed in DNA nanotechnology.7-12

The DNA origami method made possible the construction of

† iCeMS, Kyoto University.‡ Department of Chemistry, Kyoto University.§ CREST, JST.

(1) Halford, S. E.; Marko, J. F. Nucleic Acids Res. 2004, 32, 3040.(2) Jeltsch, A. ChemBioChem 2002, 3, 274.(3) Cheng, X. D. Annu. ReV. Biophys. Biomol. Struct. 1995, 24, 293.

(4) Youngblood, B.; Reich, N. O. J. Biol. Chem. 2006, 281, 26821.(5) Yun, C. S.; Khitrov, G. A.; Vergona, D. E.; Reich, N. O.; Strouse,

G. F. J. Am. Chem. Soc. 2002, 124, 7644.(6) Wolfe, S. A.; Ferentz, A. E.; Grantcharova, V.; Churchill, M. E.;

Verdine, G. L. Chem. Biol. 1995, 2, 213.(7) Seeman, N. C. Nature 2003, 421, 427.(8) Feldkamp, U.; Niemeyer, C. M. Angew. Chem., Int. Ed. 2006, 45,

1856.

Published on Web 01/15/2010

10.1021/ja907649w 2010 American Chemical Society1592 9 J. AM. CHEM. SOC. 2010, 132, 1592–1597

addressable two-dimensional (2D) and three-dimensional (3D)scaffolds for the production of complicated patterns and forplacingandarrangingfunctionalmoleculesandnanomaterials.13-16

Selective positioning of the functional molecules and nanopar-ticles has been achieved on the DNA origami scaffold system.17

The method is valuable for the preparation of predesigned 2DDNA structures, which leads to the defined assembly of middle-sized nanostructures.

Here we report the creation of two different sets of doublehelical tensions, tense and relaxed states of double helices, to

examine the effect on DNA methylation by M.EcoRI using thesestructurally controlled duplex DNAs. To prepare the two duplexDNA states, we designed and prepared the 2D DNA nanostruc-ture, named the “DNA frame,” using the DNA origami method.A DNA frame (80 nm × 90 nm) is a rectangle with a vacantrectangle area (40 nm × 40 nm) inside, in which four connectionsites were introduced for hybridization of the two differentdouble-strand DNAs (Figure 1A). The length of the doublestrands fitted in the DNA frame is 64mer double-strand DNA,which corresponds to the same length of the duplex parallel tothe frame (Figure 1B). We also introduced a 74mer double-strand DNA, which can allow bending by 60° at the targetsequence. We introduced 64mer and 74mer double strands intothe frame, meaning that the 64mer double-strand DNA just fittedas a tense strand in the frame, while the 74mer double strand isa more relaxed duplex that can allow bending of the double-strand DNAs. Using these tension-controlled double strands inthe DNA frame, we intended to control the DNA methylationand examine the effect of the DNA bending on the methylationreaction.

Results and Discussion

DNA Frame Formation and Incorporation of Double-StrandDNAs. DNA frame formation was carried out using M13mp18single-stranded DNA and sequence-designed DNA strands(staple strands) in a buffer containing Tris buffer (pH 7.6),EDTA, and Mg2+. The mixtures were annealed from 85 to 15°C by decreasing the temperature at a rate of 1.0 °C/min. Afterannealing, the DNA structures were observed by a fast-scanningatomic force microscope (AFM) in the same buffer solution.

AFM images of DNA frame structures are shown in Figure2. After self-assembly by annealing, the predesigned shape ofDNA frame structure presented in Figure 1A was observed(Figure 2A, B), and all the imaged structures maintained the

(9) Niemeyer, C. M., Mirkin, C. A., Eds. Nanobiotechnology: Concepts,Applications and PerspectiVes; Wiley-VCH Verlag GmbH & Co.KGaA: Weinheim, Germany, 2004.

(10) Aldaye, F. A.; Palmer, A. L.; Sleiman, H. F. Science 2008, 321, 1795.(11) Endo, M.; Sugiyama, H. ChemBioChem. 2009, 10, 2420.(12) (a) Shih, W. M.; Quispe, J. D.; Joyce, G. F. Nature 2004, 427, 618.

(b) Goodman, P.; Schaap, I. A.; Tardin, C. F.; Erben, C. M.; Berry,R. M.; Schmidt, C. F.; Turberfield, A. J. Science 2005, 310, 1661. (c)He, Y.; Ye, T.; Su, M.; Zhang, C.; Ribbe, A. E.; Jiang, W.; Mao, C.Nature 2008, 452, 198.

(13) Rothemund, P. W. Nature 2006, 440, 297.(14) (a) Andersen, E. S.; Dong, M.; Nielsen, M. M.; Jahn, K.; Lind-

Thomsen, A.; Mamdouh, W.; Gothelf, K. V.; Besenbacher, F.; Kjems,J. ACS Nano 2008, 2, 1213. (b) Kuzuya, A.; Kimura, M.; Numajiri,K.; Koshi, N.; Ohnishi, T.; Okada, F.; Komiyama, M. ChemBioChem.2009, 10, 1811.

(15) (a) Douglas, S. M.; Chou, J. J.; Shih, W. M. Proc. Natl. Acad. Sci.U.S.A. 2007, 104, 6644. (b) Andersen, E. S.; Dong, M.; Nielsen, M. M.;Jahn, K.; Subramani, R.; Mamdouh, W.; Golas, M. M.; Sander, B.;Stark, H.; Oliveira, C. L.; Pedersen, J. S.; Birkedal, V.; Besenbacher,F.; Gothelf, K. V.; Kjems, J. Nature 2009, 459, 73. (c) Kuzuya, A.;Komiyama, M. Chem. Commun. 2009, 4182. (d) Ke, Y.; Sharma, J.;Liu, M.; Jahn, K.; Liu, Y.; Yan, H. Nano Lett. 2009, 9, 2445. (e)Endo, M.; Hidaka, K.; Kato, T.; Namba, K.; Sugiyama, H. J. Am.Chem. Soc. 2009, 131, 15570.

(16) (a) Douglas, S. M.; Dietz, H.; Liedl, T.; Hogberg, B.; Graf, F.; Shih,W. M. Nature 2009, 459, 414. (b) Dietz, H.; Douglas, S. M.; Shih,W. M. Science 2009, 325, 725. (c) Ke, Y.; Douglas, S. M.; Liu, M.;Sharma, J.; Cheng, A.; Leung, A.; Liu, Y.; Shih, W. M.; Yan, H.J. Am. Chem. Soc. 2009, 131, 15903.

Figure 1. DNA frame structure and the incorporation of double-strand DNAs into the DNA frame. (A) The dimensions of the DNA frame structure withthe vacant area inside the frame and the four connection sites A, B, C, and D. (B) Scheme for the incorporation of two different double-strand DNAs byself-assembly using the connection sites.

J. AM. CHEM. SOC. 9 VOL. 132, NO. 5, 2010 1593

Regulation of DNA Methylation A R T I C L E S

designed shapes. Most of the DNA frames further assembledas extended structures, in which multiple frames connected byπ-stacking interactions. The size of the outside edges of theDNA frame structure were 107 ( 5 nm × 101 ( 4 nm, whichare a little bit larger than the expected size (90 nm × 80 nm).Four connection sites were observed in the inside vacant area,and the dimensions were 45 ( 3 nm × 43 ( 3 nm, which arealso slightly larger than the original design (40 nm × 40 nm).The positions of the four connection points were identified bythe absence of the corner of the right bottom side of the DNAframe (orientation marker).

To introduce further double-strand DNAs into the DNAframe, we assembled the two different double-strands. A tense64mer double-strand and a relaxed 74mer double-strand wereintroduced between the connection sites A-B and C-D,respectively. To selectively hybridize these strands into theconnection points, we introduced corresponding 16mer singlestrands to these strands, which have an overhang at both 5′-sides (Table S1). Hybridization of the two duplexes wasachieved in a solution containing the DNA frame and twoduplexes (2 equiv) by heating at 40 °C and then cooling to 15°C at a rate of 1.0 °C/min. After hybridization, AFM imageswere obtained in the same buffer solution. Using this hybridiza-tion condition, two double-strand DNAs were incorporated intothe DNA frame, and 95% of the frames were observed as atwo-duplex attached form and 5% as a one-duplex attached form(113 frames analyzed) (Figure 2C, D). These results indicate

that the two duplexes are long enough to fit into these connectionsites and are easily incorporated into the frame structure bysequence-selective hybridization.

Fast-Scanning AFM Imaging of Double-Strand DNAs andM.EcoRI Binding. Detailed analysis of the tense 64mer double-strand and relaxed 74mer double-strand DNAs were performedusing fast-scanning AFM imaging, which can successivelyacquire one AFM image per second.18 Dynamic movement ofthe duplexes was observed by successive scanning. Images ofa typical strand movement in both 64mer and 74mer double-strand DNAs are shown in Figure 3A. Dynamic movement ofthe 74mer double strand was observed in the DNA frame, whilethe movement of the 64mer double strand was modest. Thisresult indicates that the 74mer double strand had more freedomof movement compared with the 64mer double strand. Fromthe trajectory of the 74mer double-strand DNA, the center ofthe duplex moved in a radius of <7 nm. The results also indicatethat the 74mer double-strand can bend ∼60° in the DNA framenanostructure.

The binding of M.EcoRI to the double-strand DNAs in theDNA frame was observed by AFM. The DNA frame was placedon the mica surface in the buffer, the mica plate was rinsed bythe buffer to remove excess staples and specific double strands,and then M.EcoRI solution was applied onto the mica plate.The binding of M.EcoRI to the double strand in the DNA framewas observed, and the height of the M.EcoRI binding site was3.6 nm, which was 2.5 nm higher than that of the double strands(1.1 nm) (Figures 3B, 3C, and S2). The binding of M.EcoRI tothe duplexes was 26% of the DNA frame, which is roughlyconsistent with the reported Kd value (43 nM) of M.EcoRI atthe concentration of 10 nM DNA frame (108 frames analyzed)

(17) (a) Chhabra, R.; Sharma, J.; Ke, Y.; Liu, Y.; Rinker, S.; Lindsay, S.;Yan, H. J. Am. Chem. Soc. 2007, 129, 10304. (b) Ke, Y.; Lindsay, S.;Chang, Y.; Liu, Y.; Yan, H. Science 2008, 319, 180. (c) Sharma, J.;Chhabra, R.; Andersen, C. S.; Gothelf, K. V.; Yan, H.; Liu, Y. J. Am.Chem. Soc. 2008, 130, 7820. (d) Rinker, S.; Ke, Y.; Liu, Y.; Chhabra,R.; Yan, H. Nat. Nanotechnol. 2008, 3, 418. (e) Shen, W.; Zhong, H.;Neff, D.; Norton, M. L. J. Am. Chem. Soc. 2009, 131, 6660. (f) Gu,H.; Chao, J.; Xiao, S.; Seeman, N. C. Nat. Nanotechnol. 2009, 4, 245.

(18) Ando, T.; Kodera, N.; Takai, E.; Maruyama, D.; Saito, K.; Toda, A.Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 12468.

Figure 2. AFM images of the DNA frame and the DNA frame having two double-strand DNAs. (A, B) AFM image of self-assembled DNA frame and theexpanded image. (C, D) AFM image of DNA frame carrying two different double-strand DNAs and the expanded image. A triangle in the figure shows aright-bottom lacking corner of the DNA frame (orientation marker). Image size: (A), (C) 1000 nm × 750 nm; (B), (D) 300 nm × 225 nm.

1594 J. AM. CHEM. SOC. 9 VOL. 132, NO. 5, 2010

A R T I C L E S Endo et al.

(Figure S3).19 The ratios of the M.EcoRI binding for the 64merand 74mer double strands were 13% and 87%, respectively,indicating the preferential binding to the 74mer double strand(Figures 3B, 3C, and S3). The reason for these observations isthat the 74mer double strand may more easily accommodatethe structural change induced by the binding of M.EcoRI tothe duplex.

M.EcoRI binding to the DNA strand was directly observedby fast-scanning AFM. Successive images were obtained as oneimage for 1 s. When M.EcoRI bound to the 74mer double strand,the movement of the single M.EcoRI was observed in the DNAframe (Figure 3C). The distance of the movement from thecenter of the connectors C and D in the frame was <26 nm,indicating that the 74mer double strand is flexible enough forM.EcoRI to smoothly move on the duplex DNA. On the otherhand, when M.EcoRI bound to the 64mer double strand, themovement of the M.EcoRI was relatively small (Figure 3B).The distance of the movement of M.EcoRI on the 64mer doublestrand from the center of the connectors A and B in the frame

was <6 nm. These results indicate that the 74mer double strandis flexible enough to change the duplex structure, unlike the64mer double-strand, when M.EcoRI bound to the duplexes inthe DNA frame.

Effect of Tensions of Double-Strand DNAs in the DNAFrame on DNA Methylation. We next examined the M.EcoRImethylation and R.EcoRI cleavage to identify DNA methylationin the DNA frame by AFM imaging. Methylation of the tense64mer double-strand and relaxed 74mer double-strand DNAsin the DNA frame was performed by treatment with M.EcoRIin the presence of S-adenosyl-L-methionine (SAM). Aftersuccessive treatment with R.EcoRI, the methylation was ana-lyzed by observing the preserved double-strand DNAs by AFM.When the target sequence was methylated, the sequence is notcleaved by R.EcoRI, indicating that the DNA strand in the DNAframe is preserved. On the other hand, when the target sequenceis not methylated, the double-strand DNA cleaved by R.EcoRIshould have remained in the DNA frame. The DNA frames aftertreatment with enzymes were analyzed by AFM (Figure 4). First,R.EcoRI completely cleaved both DNA strands in the frameunder our experimental conditions (Figure S4). Although(19) Reich, N. O.; Dabzitz, M. J. Biochemistry 1992, 31, 1937.

Figure 3. Successive fast-scanning AFM images of two double-strand DNAs in the DNA frame and M.EcoRI binding to the double strands in the DNAframe. One image was obtained every 1 s. (A) 64mer and 74mer double strands in the DNA frame. (B, C) Single M.EcoRI binding to the 64mer and 74merdouble strands in the DNA frame, respectively. An arrow in the figure indicates M.EcoRI. A triangle in the figure shows a right-bottom lacking corner ofthe DNA frame (orientation marker), and 64mer and 74mer represent the 64mer and 74mer double-strand DNAs, respectively. The numbers in the figuresare the lapsed times (s) of the successive images. Image size 225 nm × 225 nm.

J. AM. CHEM. SOC. 9 VOL. 132, NO. 5, 2010 1595

Regulation of DNA Methylation A R T I C L E S

R.EcoRI needs unwinding of the duplex by 25° for binding,20

both strands may unwind like a usual duplex. Second, aftertreatment with M.EcoRI, no cleavage of the duplexes in theDNA frames was observed (Figure S5). Finally, after reactionwith M.EcoRI and subsequent treatment with R.EcoRI, 43%of the 74mer double strands and 13% of the 64mer doublestrands were preserved (Figure 4), in which 118 frames wereanalyzed. The results show that the 74mer double strand wasnot as effectively cleaved as the 64mer double strand and therelaxed 74mer duplex is a better substrate for methylation ofM.EcoRI than the tense 64mer duplex.

Biochemical Analysis of the Methylation in the DNAFrame by Real-Time PCR. To examine the effect of the tensionof the double-strand DNAs on methylation, we performed thebiochemical analysis of the methylation in the DNA frame aftergel purification. The tense 64mer and relaxed 74mer doublestrands in the DNA frame were treated with M.EcoRI. Thenmethylation was analyzed by cleavage with R.EcoRI, and theinitial concentration of the preserved duplex DNAs wasestimated by real-time PCR (Figure 5). The scheme for theanalysis of methylation is shown in Figure 5A. In our strategy,when the target sequence is methylated, the sequence is notcleaved by R.EcoRI (Figure S5), meaning that the full-sizedincorporated duplex DNAs should be amplified by PCR usingtwo specific primers. On the other hand, when the targetsequence is not methylated, cleavage of the target duplex DNAsby R.EcoRI should result in no amplification of the target duplex

DNA by PCR. For this experiment, we employed a gel-purifiedDNA frame carrying two duplexes to avoid background PCRamplification due to the contamination of the free target doublestrands (Figure S6). When the purified DNA frame carryingtwo duplexes were amplified by using the specific primers for64mer and 74mer strands, target PCR products were obtainedfrom both combinations of primers (Figure S7). When theduplex DNAs were treated with M.EcoRI in the absence of theDNA frame, target PCR products were obtained from methy-lated duplexes, indicating that methylation did not reduce thePCR amplification (Figure S8).

To estimate the initial concentration of the target DNA inthe DNA frame by real-time PCR, we performed controlexperiments for comparison in the presence and absence ofM.EcoRI and R.EcoRI. In the control reactions using 64merand 74mer double strands in the DNA frame without M.EcoRI,both strands were amplified at initial concentrations of 0.34-0.44nM. On the other hand, after treatment of these strands withR.EcoRI, the initial concentration dropped to 0.04-0.10 nM.These results indicate that the R.EcoRI cleavage worked toreduce the PCR amplification, which was comparable to thelow initial concentration of target double strands. After treatmentwith M.EcoRI, if the target sequence is methylated, subsequentR.EcoRI cleavage should be prohibited, resulting in the pre-served concentration of target double strands. In the case ofthe 64mer double strand treated with M.EcoRI, the initialconcentration of the corresponding strand after R.EcoRI cleav-age dropped from 0.38 to 0.10 nM. On the other hand, whenthe 74mer double strand was treated with M.EcoRI and further

(20) McClarin, J. A.; Frederick, C. A.; Wang, B.; Greene, P.; Boyer, H. W.;Grable, J.; Rosenberg, J. M. Science 1986, 234, 1526.

Figure 4. AFM images of DNA double-strand cleavage in the DNA frame after treatment with M.EcoRI followed by R.EcoRI. Wide image (A) andexpanded images (B) of the reaction products. A triangle represents an orientation marker in the DNA frame, and 74mer represents the 74mer duplex DNA.Image sizes: (A) 1000 nm × 750 nm; (B) 500 nm × 375 nm.

Figure 5. Quantification of the methylation reaction with M.EcoRI in the DNA frame. (A) Scheme for the method for the biochemical analysis of themethylation. Two DNA strands were treated with M.EcoRI in the first step and then treated with R.EcoRI in the second step. Finally, the specific sequenceof the samples was quantified by real-time PCR. (B) The results of real-time PCR for the estimation of the initial concentration of the 64mer and 74merdouble strands in the DNA frame after treatment with enzymes. Without M.EcoRI treatment, R.EcoRI cleavage gave a clear difference in the results (comparethe two left bars). The data were collected from three independent experiments. Primers used for the real-time PCR are listed in Table S2.

1596 J. AM. CHEM. SOC. 9 VOL. 132, NO. 5, 2010

A R T I C L E S Endo et al.

cleaved with R.EcoRI, the initial concentration remainedunchanged at 0.25 nM with and without R.EcoRI cleavage.These results indicate that the 74mer double strand in the framewas strongly methylated under this reaction condition.

Bending of the duplex DNA for methyl transfer to the N6-position of the specific adenine is required for the reaction toproceed.4,5 In our DNA frame system, tensed and relaxed doublestrands can be created by using the defined length in thenanostructural DNA scaffold. The relaxed 74mer double strandcan accommodate M.EcoRI to bind and bend the targetsequence. On the other hand, the tense 64mer double strandallows binding of the M.EcoRI, but this strand is a poor substratefor bending, resulting in poor methylation of the target sequence.

Conclusion

We have designed and prepared a DNA frame nanostructureusing the DNA origami method and created two different doublehelical tensions in the frame structure to control and examinethe effect of the tensions on DNA methylation of EcoRImethyltransferase. The methylation reaction can be controlledusing two structurally fixed types of double-strand DNAsconstructed in the DNA frame nanostructure. We have shownthe importance of DNA strand relaxation in allowing doublehelix bending during the methylation reaction of M.EcoRI. Tothe best of our knowledge this is the first demonstration of therole of DNA flexibility on an enzymatic modification reaction.This method can be used for other DNA-modifying enzymesand DNA-repair enzymes that bend the double helix during theenzymatic reaction. In addition, the DNA frame structure couldbe a versatile scaffold to observe the single-molecule behaviorof DNA-binding enzymes, including binding to DNA, searchingthe specific sequence, catalyzing the reaction substrate, anddissociation from the DNA. The method can be extended tothe direct observation of various enzymatic phenomena in thedesigned nanoscale space.

Materials and Methods

Design of DNA Frame Structure. The DNA sequence designof the DNA frame structure followed the rules of the DNA origamimethod. The sequence of the M13mp18 single-stranded DNA wasused, and the staple strands (most of them are 32mer) were assignedfor the formation of the designed DNA frame. The DNA framelacks the right bottom corner for identification of the orientationof the frame (orientation marker). The sequences and positions ofthe staple strands are listed in the Supporting Information.

DNA Frame Formation and Introduction of the Double-Strand DNAs. The DNA frame was assembled in a 20 µL solutioncontaining M13mp18 single-stranded DNA (New England Biolabs)10 nM, staple strands (226 strands) 50 nM (5 equiv), 20 mM Trisbuffer (pH 7.6), 1 mM EDTA, and 10 mM MgCl2. The mixturewas annealed from 95 to 15 °C at a rate of -1.0 °C/min.Hybridization of the two duplexes was achieved in the solutioncontaining a DNA frame and two duplexes (2 equiv) by heating at50 °C and then cooling to 15 °C at a rate of 1.0 °C/min using a

thermal cycler. The sequences of the staple strands are listed inthe Supporting Information.

Fast-Scanning AFM Imaging of the DNA Frame. AFM imageswere obtained on a fast-scanning AFM system (Nano Live Vision,RIBM, Tsukuba, Japan) using a silicon nitride cantilever (OlympusBL-AC10EGS). The sample (2 µL) was adsorbed on a freshlycleaved mica plate for 5 min at room temperature and then washedthree times with the same buffer solution. Scanning was performedin the same buffer solution.

Observation of Binding of the M.EcoRI to the TargetDouble-Strand DNAs in the DNA Frame. The binding of theM.EcoRI to the duplexes in the DNA frame was examined in asolution containing EcoRI methyltransferase (2 units, New EnglandBiolabs), 50 mM Tris-HCl (pH 7.5), 100 mM NaCl, 10 mM MgCl2,and 1 mM DTT. The solution mixtures (2 µL) were added ontothe DNA frame with two double strands adsorbed on the mica plate,and the mica plate was washed three times using the same bufferto remove the excess M.EcoRI. Scanning was performed in thesame buffer solution.

Methylation of the Target Double-Strand DNAs in theDNA Frame. A DNA frame having both 64mer and 74mer doublestrands was purified by 1.0% agarose gel electrophoresis, and thecorresponding band was cut and recovered using a Freeze ’NSqueeze DNA gel extraction column (BioRad) by following themanufacture’s protocol (Figure S5). The methylation reaction wasperformed in a 10 µL solution containing the gel-purified DNAframe with two double strands, M.EcoRI, 50 mM Tris-HCl (pH7.5), 100 mM NaCl, 10 mM MgCl2, 1 mM DTT, 0.08 mM SAMat 37 °C for 1.5 h. Then R.EcoRI (2 units) was added to the reactionmixture, and the mixture was incubated at 37 °C for 1.5 h.

Quantification of Methylation in the DNA Frame byReal-Time PCR. Real-time PCR and subsequent calculations wereperformed with a 7300 Real-Time PCR System (Applied Biosys-tems), which detects the signal emitted from fluorogenic probesduring PCR. Real-time PCR was performed with FastStart UniversalSYBR Green Master (Roche). The PCR mixture contained 1×FastStart Universal SYBR Green Master, 300 nM forward andreverse primers, and the gel-purified DNA frame carrying twodouble strands in a total volume of 25 µL. Each 40 PCR cycle wasprogrammed as a 15 s denaturation step at 95 °C and 1 minhybridization step, with probes and primers and target DNA strandsat 58 °C. The standard curve was obtained using 500, 100, 20, 4,and 0.8 pM DNA templates. The incorporated double-strand DNAsand primer sequences are listed in Table S2.

Acknowledgment. This work was supported by Core Researchfor Evolutional Science and Technology (CREST) of JST, a Grant-in-Aid for Science Research from the MEXT, Japan, and TokuyamaScience Foundation (M.E.). The authors acknowledge the supportby the Global COE Program “Integrated Materials Science” (#B-09).

Supporting Information Available: Experimental details. Thismaterial is available free of charge via the Internet at http://pubs.acs.org.

JA907649W

J. AM. CHEM. SOC. 9 VOL. 132, NO. 5, 2010 1597

Regulation of DNA Methylation A R T I C L E S